In theory, virtually any viral protein could be expressed by recombinant means and used as a vaccine. This could be restricted to just those proteins containing neutralizing epitopes, which are the target of neutralizing antibodies. In addition, it also depends on immunogenicity, since vaccine potency will vary from one protein to another. This has become a major problem for vaccines against HIV-1, since envelope glycoprotein gp120 has important neutralizing epitopes, but is not very immunogenic in man. Our approach was to prepare vaccine conjugates, by coupling gp120 to a variety of immunogenic proteins. These should act as carrier proteins, conferring part of their own immunogenicity on gp120. To do this, we have developed a crosslinking strategy, based on the oligosaccharide side chains of gp120,which has 22 per mole. They are oxidized with neuraminidase/galactose oxidase, which produces reactive aldehydes, while retaining the native protein conformation. We then developed a new family of crosslinkers, which contain hydrazide at one end and a free SH at the other. This reacts with the aldehydes and produces free SH groups. In the case of gp120, we could introduce up to 5 free SH groups per mole. This was enough to allow crosslinking with a variety of maleimide modified proteins, including myoglobin and tetanus toxoid. The tetanus toxoid conjugate will be analyzed further for immunogenicity by priming with tetanus and boosting with the conjugate. It will also be analyzed in small monkeys, where monomeric gp120 is not very immunogenic. Future studies include using mutated tetanus toxin, which has many more NH2 groups available for coupling. The resulting antibodies will be tested for neutralizing activity on the same and divergent HIV-1 isolates, as well as on fresh isolates that have recently transmitted infection.